A method of organizing computer resources includes receiving a specification defining a plurality of quiescence groups of independent component instances for each of at least two services, and performing a first load balancing of the quiescence groups across a plurality of physical servers to define a plurality of supergroups while assigning each of the physical servers across the supergroups.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
2. The method of claim 1, wherein the placement of the quiescence groups into the supergroups uses a minimum number of the physical servers.
This invention relates to optimizing the placement of quiescence groups within a computing system to minimize the number of physical servers required. Quiescence groups are collections of virtual machines or other workloads that must be temporarily paused or frozen for maintenance, backups, or other operations. The challenge addressed is efficiently organizing these groups into larger supergroups while reducing the physical server footprint, thereby improving resource utilization and operational efficiency. The method involves analyzing the dependencies and requirements of multiple quiescence groups to determine the most efficient way to consolidate them into supergroups. By strategically assigning groups to supergroups, the system ensures that the minimum number of physical servers is used while still meeting all operational constraints. This approach reduces hardware costs, power consumption, and administrative overhead. The solution is particularly useful in large-scale data centers or cloud environments where resource optimization is critical. The method may also include dynamic adjustments based on real-time workload demands or changes in the system configuration.
3. The method of claim 1, wherein anti-collocation of dependent ones of the component instances to the physical servers is enforced by the placement of the quiescence groups into the supergroups.
This invention relates to distributed computing systems, specifically methods for managing the placement of component instances across physical servers to improve fault tolerance and resource utilization. The problem addressed is ensuring that dependent component instances are not co-located on the same physical server, which could lead to cascading failures if that server experiences issues. The solution involves organizing component instances into quiescence groups and then assigning these groups to supergroups, where the supergroups enforce anti-collocation rules. Quiescence groups represent sets of component instances that can be safely paused or restarted together, while supergroups act as containers that distribute these groups across multiple physical servers. By placing quiescence groups into supergroups, the system ensures that dependent instances are spread across different servers, reducing the risk of correlated failures. The method also optimizes resource allocation by dynamically adjusting group assignments based on workload demands and server availability. This approach improves system reliability and resilience while maintaining efficient resource usage.
4. The method of claim 1, further comprising assigning the physical servers such that no two supergroups have component instances coexisting on a same one of the physical servers.
This invention relates to distributed computing systems, specifically methods for managing physical servers in a cloud computing environment to enhance security and resource isolation. The problem addressed is the risk of security breaches or performance degradation when multiple virtualized computing instances (component instances) from different logical groups (supergroups) share the same physical server, potentially allowing unauthorized access or resource contention. The method involves assigning physical servers to component instances in a way that ensures no two supergroups have their instances running on the same physical server. This strict isolation prevents cross-group interference, reducing the risk of data leaks, unauthorized access, or performance bottlenecks caused by shared resources. The approach likely involves dynamic allocation algorithms that consider the supergroup affiliations of component instances when distributing them across available physical servers, ensuring compliance with the isolation requirement while optimizing resource utilization. The method may also include monitoring and rebalancing server assignments to maintain isolation as workloads change, ensuring continuous adherence to the security and performance constraints. This technique is particularly useful in multi-tenant cloud environments where multiple independent organizations or applications share the same infrastructure but require strict logical separation.
5. The method of claim 1, further comprising performing a maintenance function on portions of the computer system supporting sequential ones of the supergroups without interrupting the services.
This invention relates to maintaining a computer system while ensuring continuous service availability. The system is organized into multiple supergroups, each containing multiple groups, which in turn contain multiple nodes. The method involves performing maintenance functions on portions of the computer system that support sequential supergroups without interrupting services. This allows maintenance tasks to be carried out on different parts of the system in a staggered manner, ensuring that at least some supergroups remain operational at all times. The maintenance functions may include software updates, hardware repairs, or other system maintenance tasks. By sequentially targeting different supergroups, the system avoids downtime while still allowing comprehensive maintenance to be performed. The method ensures that services remain uninterrupted by carefully coordinating maintenance activities across the supergroups, preventing any single maintenance operation from affecting the entire system. This approach is particularly useful in large-scale distributed systems where continuous availability is critical.
6. The method of claim 5, wherein performing the maintenance function includes taking down, simultaneously, the independent component instances of the quiescence groups within a given supergroup.
This invention relates to systems for managing and maintaining distributed computing environments, particularly in scenarios where multiple independent component instances must be taken down for maintenance without disrupting overall system functionality. The problem addressed is the challenge of performing maintenance on distributed systems where components are grouped into quiescence groups and supergroups, requiring coordinated shutdowns to avoid service interruptions. The method involves performing a maintenance function on a distributed system where components are organized into quiescence groups and supergroups. Each quiescence group contains independent component instances that can be taken down without affecting other groups. A supergroup is a collection of these quiescence groups, allowing for hierarchical management. The maintenance function includes the step of taking down all independent component instances within the quiescence groups of a given supergroup simultaneously. This simultaneous shutdown ensures that maintenance is performed efficiently while maintaining system stability. The method may also involve verifying that the shutdown process is completed successfully before proceeding with further maintenance tasks. This approach minimizes downtime and ensures that maintenance operations do not disrupt the overall system's availability.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
September 30, 2021
May 7, 2024
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.